journal article·2022-10-31Open access

Fall Detection System Based on Simple Threshold Method and Long Short-Term Memory: Comparison with Hidden Markov Model and Extraction of Optimal Parameters

Seung Su Jeong, Nam Ho Kim, Yun Seop Yu

Applied Sciences· Volume 12· Issue 21

Abstract

In an aging global society, a few complex problems have been occurring due to falls among the increasing elderly population. Therefore, falls are detected using a pendant-type sensor that can be worn comfortably for fall detection. The sensed data are processed by the embedded environment and classified by a long-term memory (LSTM). A fall detection system that combines a simple threshold method (STM) and LSTM, the STM-LSTM-based fall detection system, is introduced. In terms of training data accuracy, the proposed STM-LSTM-based fall detection system is compared with the previously reported STM-hidden Markov model (HMM)-based fall detection system. The training accuracy of the STM-LSTM fall detection system is 100%, while the highest training accuracy by the STM-HMM-based one is 99.5%, which is 0.5% less than the best of the STM-LSTM-based system. In addition, in the optimized LSTM fall detection system, this may be overfitted because all data are trained without separating any validation data. In order to resolve the possible overfitting issue, training and validation data are evaluated separately in 4:1, and then in terms of validation data accuracy of the STM-LSTM-based fall detection system, optimal values of the parameters in LSTM and normalization method are found as follows: best accuracy of 98.21% at no-normalization, no-sampling, 128hidden layer nodes, and regularization rate of 0.015. It is also observed that as the number of hidden layer nodes or sampling interval increases, the regularization rate at the highest value of accuracy increases. This means that overfitting can be suppressed by increasing the regularization, and thus an appropriate number of hidden layer nodes and a regularization rate must be selected to improve the fall detection efficiency.
Source: crossref · Quality: baseline

Research context

OpenAlex topics: Context-Aware Activity Recognition Systems · IoT and Edge/Fog Computing · IoT-based Smart Home Systems

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

Affiliations

ICT & Robotics Engineering, Semiconductor Convergence Engineering, AISPC Laboratory, IITC, Hankyong National University, 327 Jungang-ro, Anseong-si 17579, Gyenggi-do, Korea

Department of Embedded System, Bundang Convergence Technology Campus of Korea Polytechnic, 5, Hwangsaeul-ro 329 beon-gil, Sengnam 13590, Gyeonggi-do, Korea

Funding

Ministry of Education

References (48)

Show source-provided references
1. Falls among elderly and its relation with their health problems and surrounding environmental factors in Riyadh10.4103/jfcm.JFCM_48_17 ↗
2. (2020, March 09). World Health Organization. Available online: https://www.who.int/zh/news-room/fact-sheets/detail/falls.
3. (2020, March 09). National Health Administration, Ministry of Health and Welfare, Available online: https://www.hpa.gov.tw/Pages/Detail.aspx?nodeid=807&pid=4326.
4. Challenges, issues and trends in fall detection systems10.1186/1475-925X-12-66 ↗
5. Elderly fall detection systems: A literature survey10.1364/FIO.2020.FM7D.1 ↗
6. A survey on recent advances in wearable fall detection systems10.1155/2020/2167160 ↗
7. Research of fall detection and fall prevention technologies: A systematic review10.1109/ACCESS.2019.2922708 ↗
8. Taylor, R.M., Marc, E.C., Vangeli, S.M., Anne, H.H.N., and Coralys, C.R. (2018). SmartFall: A smartwatch-based fall detection system using deep learning. Sensors, 18.10.3390/s18103363 ↗
9. Vilarinho, T., Farshchian, B., Bajer, D.G., Dahl, O.H., Egge, I., Hegdal, S.S., Lones, A., and Slettevold, J.N. (2015, January 26–28). A combined smartphone and smartwatch fall detection system. Proceedings of the 2015 IEEE International Conference on Computer and Information Technology, Ubiquitous Computing and Communications, Dependable, Autonomic and Secure Computing, Pervasive Intelligence and Computing, Liverpool, UK.10.1109/CIT/IUCC/DASC/PICOM.2015.216 ↗
10. Casilari, E., and Oviedo-Jiménez, M.A. (2015). Automatic fall detection system based on the combined use of a smartphone and a smartwatch. PLoS ONE, 10.10.1371/journal.pone.0140929 ↗
11. Habib, M.A., Mohktar, M.S., Kamaruzzaman, S.B., Lim, K.S., Pin, T.M., and Ibrahim, F. (2014). Smartphone-based solutions for fall detection and prevention: Challenges and open issues. Sensors, 14.10.3390/s140407181 ↗
12. Rakhman, A.Z., and Nugroho, L.E. (2014, January 8). Fall detection system using accelerometer and gyroscope based on smartphone. Proceedings of the 1st International Conference on Information Technology, Computer, and Electrical Engineering, Semarang, Indonesia.10.1109/ICITACEE.2014.7065722 ↗
13. Yavu, G., Kocak, M., Ergun, G., Alemdar, H.O., Yalcin, H., Incel, O.D., and Ersoy, C. (2010, January 2). A smartphone based fall detector with online location support. Proceedings of the International Workshop on Sensing for App Phones (ACM), Zurich, Switzerland.
14. He, Y., Li, Y., and Bao, S.D. (2012, January 5–7). Fall detection by built-in tri-accelerometer of smartphone. Proceedings of the 2012 IEEE-EMBS International Conference on Biomedical and Health Informatics, Hong Kong, China.
15. A smartphone-based fall detection system
16. Emergency-monitoring system based on newly-developed fall detection algorithm
17. Fall-detection algorithm using 3-axis acceleration: Combination with simple threshold and hidden Markov model10.1155/2014/896030 ↗
18. Jiang, M., Chen, Y., Zhao, Y., and Cai, A. (2013, January 17–20). A real-time fall detection system based on HMM and RVM. Proceedings of the 2013 Visual Communications and Image Processing (VCIP), Kuching, Malaysia.10.1109/VCIP.2013.6706385 ↗
19. A real-time, multiview fall detection system: A LHMM-based approach10.1109/TCSVT.2008.2005606 ↗
20. Decision tree analysis of construction fall accidents involving roofers10.1016/j.eswa.2014.10.009 ↗
21. A fall detection system using k-nearest neighbor classifier10.1016/j.eswa.2010.04.014 ↗
22. Liu, L., Popescu, M., Skubic, M., Rantz, M., Yardibi, T., and Cuddihy, P. (2011, January 23–26). Automatic fall detection based on Doppler radar motion signature. Proceedings of the 2011 5th International Conference on Pervasive Computing Technologies for Healthcare (PervasiveHealth), Dublin, Ireland.
23. Zhang, T., Wang, J., Xu, L., and Liu, P. (2006). Fall Detection by Wearable Sensor and One-Class SVM Algorithm. Intelligent Computing in Signal Processing and Pattern Recognition, Springer.10.1007/978-3-540-37258-5_104 ↗
24. An introduction to hidden Markov models and Bayesian networks10.1142/S0218001401000836 ↗
25. Anguita, D., Ghio, A., Greco, N., Oneto, L., and Ridella, S. (2010, January 18–23). Model selection for support vector machines: Advantages and disadvantages of the machine learning theory. Proceedings of the 2010 International Joint Conference on Neural Networks (IJNN), Barcelona, Spain.10.1109/IJCNN.2010.5596450 ↗
26. Lin, C.B., Dong, Z., Kuan, W.K., and Huang, Y.F. (2020). A framework for fall detection based on OpenPose skeleton and LSTM/GRU models. Appl. Sci., 11.10.3390/app11010329 ↗
27. Chen, W., Jiang, Z., Guo, H., and Ni, X. (2020). Fall detection based on key points of human-skeleton using openpose. Symmetry, 12.10.3390/sym12050744 ↗
28. A real-time patient monitoring framework for fall detection10.1155/2019/9507938 ↗
29. Queralta, J.P., Gia, T.N., Tenhunen, H., and Westerlund, T. (2019, January 1–3). Edge-AI in LoRa-based health monitoring: Fall detection system with fog computing and LSTM recurrent neural networks. Proceedings of the 42nd International Conference on Telecommunications and Signal Processing (TSP), IEEE, Budapest, Hungary.10.1109/TSP.2019.8768883 ↗
30. Deep learning for fall detection: Three-dimensional CNN combined with LSTM on video kinematic data10.1109/JBHI.2018.2808281 ↗
31. Santos, G.L., Endo, P.T., Monteiro, K.H.D.C., Rocha, E.D.S., Silva, I., and Lynn, T. (2019). Accelerometer-based human fall detection using convolutional neural networks. Sensors, 19.10.3390/s19071644 ↗
32. Improving fall detection by the use of depth sensor and accelerometer10.1016/j.neucom.2015.05.061 ↗
33. A depth-based fall detection system using a Kinect® sensor10.3390/s140202756 ↗
34. Fall detection with UWB radars and CNN-LSTM architecture10.1109/JBHI.2020.3027967 ↗
35. A multimodal approach using deep learning for fall detection10.1016/j.eswa.2020.114226 ↗
36. Adhikari, K., Bouchachia, H., and Nait-Charif, H. (2017, January 8–12). Activity recognition for indoor fall detection using convolutional neural network. Proceedings of the 2017 Fifteenth IAPR International Conference on Machine Vision Applications (MVA), Nagoya, Japan.10.23919/MVA.2017.7986795 ↗
37. Casilari, E., Lora-Rivera, R., and García-Lagos, F. (2020). A study on the application of convolutional neural networks to fall detection evaluated with multiple public datasets. Sensors, 20.10.3390/s20051466 ↗
38. The vanishing gradient problem during learning recurrent neural nets and problem solutions10.1142/S0218488598000094 ↗
39. Fall recognition algorithm using gravity-weighted 3-axis accelerometer data
40. Long short-term memory
41. Zhang, Z., and Sabuncu, M. (2018, January 3–8). Generalized cross entropy loss for training deep neural networks with noisy labels. Proceedings of the 32nd Conference on Neural Information Processing Systems (NeurIPS 2018), Montréal, Canada.
42. Trimmed categorical cross-entropy for deep learning with label noise10.1049/el.2018.7980 ↗
43. Bayesian regularization of neural networks10.1007/978-1-60327-101-1_3 ↗
44. Park, P., Marco, P.D., Shin, H., and Bang, J. (2019). Fault detection and diagnosis using combined autoencoder and long short-term memory network. Sensors, 19.10.3390/s19214612 ↗
45. Li, K., Zhao, X., Bian, J., and Tan, M. (2017, January 6–9). Sequential learning for multimodal 3D human activity recognition with long-short term memory. Proceedings of the 2017 IEEE International Conference on Mechatronics and Automation (ICMA), Takamatsu, Japan.10.1109/ICMA.2017.8016048 ↗
46. Understanding and using sensitivity, specificity and predictive values10.4103/0301-4738.37595 ↗
47. A systematic review on overfitting control in shallow and deep neural networks10.1007/s10462-021-09975-1 ↗
48. Novel image classification method based on few-shot learning in monkey species